National Academies Press: OpenBook

Improved Surface Drainage of Pavements: Final Report (1998)

Chapter: Chapter 2 Literature Review and Current Practice and Techniques for Improved Surface Drainage

« Previous: Chapter 1 Introduction
Page 11
Suggested Citation:"Chapter 2 Literature Review and Current Practice and Techniques for Improved Surface Drainage." Transportation Research Board. 1998. Improved Surface Drainage of Pavements: Final Report. Washington, DC: The National Academies Press. doi: 10.17226/6357.
×
Page 11
Page 12
Suggested Citation:"Chapter 2 Literature Review and Current Practice and Techniques for Improved Surface Drainage." Transportation Research Board. 1998. Improved Surface Drainage of Pavements: Final Report. Washington, DC: The National Academies Press. doi: 10.17226/6357.
×
Page 12
Page 13
Suggested Citation:"Chapter 2 Literature Review and Current Practice and Techniques for Improved Surface Drainage." Transportation Research Board. 1998. Improved Surface Drainage of Pavements: Final Report. Washington, DC: The National Academies Press. doi: 10.17226/6357.
×
Page 13
Page 14
Suggested Citation:"Chapter 2 Literature Review and Current Practice and Techniques for Improved Surface Drainage." Transportation Research Board. 1998. Improved Surface Drainage of Pavements: Final Report. Washington, DC: The National Academies Press. doi: 10.17226/6357.
×
Page 14
Page 15
Suggested Citation:"Chapter 2 Literature Review and Current Practice and Techniques for Improved Surface Drainage." Transportation Research Board. 1998. Improved Surface Drainage of Pavements: Final Report. Washington, DC: The National Academies Press. doi: 10.17226/6357.
×
Page 15
Page 16
Suggested Citation:"Chapter 2 Literature Review and Current Practice and Techniques for Improved Surface Drainage." Transportation Research Board. 1998. Improved Surface Drainage of Pavements: Final Report. Washington, DC: The National Academies Press. doi: 10.17226/6357.
×
Page 16
Page 17
Suggested Citation:"Chapter 2 Literature Review and Current Practice and Techniques for Improved Surface Drainage." Transportation Research Board. 1998. Improved Surface Drainage of Pavements: Final Report. Washington, DC: The National Academies Press. doi: 10.17226/6357.
×
Page 17
Page 18
Suggested Citation:"Chapter 2 Literature Review and Current Practice and Techniques for Improved Surface Drainage." Transportation Research Board. 1998. Improved Surface Drainage of Pavements: Final Report. Washington, DC: The National Academies Press. doi: 10.17226/6357.
×
Page 18
Page 19
Suggested Citation:"Chapter 2 Literature Review and Current Practice and Techniques for Improved Surface Drainage." Transportation Research Board. 1998. Improved Surface Drainage of Pavements: Final Report. Washington, DC: The National Academies Press. doi: 10.17226/6357.
×
Page 19
Page 20
Suggested Citation:"Chapter 2 Literature Review and Current Practice and Techniques for Improved Surface Drainage." Transportation Research Board. 1998. Improved Surface Drainage of Pavements: Final Report. Washington, DC: The National Academies Press. doi: 10.17226/6357.
×
Page 20
Page 21
Suggested Citation:"Chapter 2 Literature Review and Current Practice and Techniques for Improved Surface Drainage." Transportation Research Board. 1998. Improved Surface Drainage of Pavements: Final Report. Washington, DC: The National Academies Press. doi: 10.17226/6357.
×
Page 21
Page 22
Suggested Citation:"Chapter 2 Literature Review and Current Practice and Techniques for Improved Surface Drainage." Transportation Research Board. 1998. Improved Surface Drainage of Pavements: Final Report. Washington, DC: The National Academies Press. doi: 10.17226/6357.
×
Page 22
Page 23
Suggested Citation:"Chapter 2 Literature Review and Current Practice and Techniques for Improved Surface Drainage." Transportation Research Board. 1998. Improved Surface Drainage of Pavements: Final Report. Washington, DC: The National Academies Press. doi: 10.17226/6357.
×
Page 23
Page 24
Suggested Citation:"Chapter 2 Literature Review and Current Practice and Techniques for Improved Surface Drainage." Transportation Research Board. 1998. Improved Surface Drainage of Pavements: Final Report. Washington, DC: The National Academies Press. doi: 10.17226/6357.
×
Page 24
Page 25
Suggested Citation:"Chapter 2 Literature Review and Current Practice and Techniques for Improved Surface Drainage." Transportation Research Board. 1998. Improved Surface Drainage of Pavements: Final Report. Washington, DC: The National Academies Press. doi: 10.17226/6357.
×
Page 25
Page 26
Suggested Citation:"Chapter 2 Literature Review and Current Practice and Techniques for Improved Surface Drainage." Transportation Research Board. 1998. Improved Surface Drainage of Pavements: Final Report. Washington, DC: The National Academies Press. doi: 10.17226/6357.
×
Page 26
Page 27
Suggested Citation:"Chapter 2 Literature Review and Current Practice and Techniques for Improved Surface Drainage." Transportation Research Board. 1998. Improved Surface Drainage of Pavements: Final Report. Washington, DC: The National Academies Press. doi: 10.17226/6357.
×
Page 27
Page 28
Suggested Citation:"Chapter 2 Literature Review and Current Practice and Techniques for Improved Surface Drainage." Transportation Research Board. 1998. Improved Surface Drainage of Pavements: Final Report. Washington, DC: The National Academies Press. doi: 10.17226/6357.
×
Page 28
Page 29
Suggested Citation:"Chapter 2 Literature Review and Current Practice and Techniques for Improved Surface Drainage." Transportation Research Board. 1998. Improved Surface Drainage of Pavements: Final Report. Washington, DC: The National Academies Press. doi: 10.17226/6357.
×
Page 29
Page 30
Suggested Citation:"Chapter 2 Literature Review and Current Practice and Techniques for Improved Surface Drainage." Transportation Research Board. 1998. Improved Surface Drainage of Pavements: Final Report. Washington, DC: The National Academies Press. doi: 10.17226/6357.
×
Page 30
Page 31
Suggested Citation:"Chapter 2 Literature Review and Current Practice and Techniques for Improved Surface Drainage." Transportation Research Board. 1998. Improved Surface Drainage of Pavements: Final Report. Washington, DC: The National Academies Press. doi: 10.17226/6357.
×
Page 31
Page 32
Suggested Citation:"Chapter 2 Literature Review and Current Practice and Techniques for Improved Surface Drainage." Transportation Research Board. 1998. Improved Surface Drainage of Pavements: Final Report. Washington, DC: The National Academies Press. doi: 10.17226/6357.
×
Page 32
Page 33
Suggested Citation:"Chapter 2 Literature Review and Current Practice and Techniques for Improved Surface Drainage." Transportation Research Board. 1998. Improved Surface Drainage of Pavements: Final Report. Washington, DC: The National Academies Press. doi: 10.17226/6357.
×
Page 33
Page 34
Suggested Citation:"Chapter 2 Literature Review and Current Practice and Techniques for Improved Surface Drainage." Transportation Research Board. 1998. Improved Surface Drainage of Pavements: Final Report. Washington, DC: The National Academies Press. doi: 10.17226/6357.
×
Page 34
Page 35
Suggested Citation:"Chapter 2 Literature Review and Current Practice and Techniques for Improved Surface Drainage." Transportation Research Board. 1998. Improved Surface Drainage of Pavements: Final Report. Washington, DC: The National Academies Press. doi: 10.17226/6357.
×
Page 35
Page 36
Suggested Citation:"Chapter 2 Literature Review and Current Practice and Techniques for Improved Surface Drainage." Transportation Research Board. 1998. Improved Surface Drainage of Pavements: Final Report. Washington, DC: The National Academies Press. doi: 10.17226/6357.
×
Page 36
Page 37
Suggested Citation:"Chapter 2 Literature Review and Current Practice and Techniques for Improved Surface Drainage." Transportation Research Board. 1998. Improved Surface Drainage of Pavements: Final Report. Washington, DC: The National Academies Press. doi: 10.17226/6357.
×
Page 37
Page 38
Suggested Citation:"Chapter 2 Literature Review and Current Practice and Techniques for Improved Surface Drainage." Transportation Research Board. 1998. Improved Surface Drainage of Pavements: Final Report. Washington, DC: The National Academies Press. doi: 10.17226/6357.
×
Page 38
Page 39
Suggested Citation:"Chapter 2 Literature Review and Current Practice and Techniques for Improved Surface Drainage." Transportation Research Board. 1998. Improved Surface Drainage of Pavements: Final Report. Washington, DC: The National Academies Press. doi: 10.17226/6357.
×
Page 39
Page 40
Suggested Citation:"Chapter 2 Literature Review and Current Practice and Techniques for Improved Surface Drainage." Transportation Research Board. 1998. Improved Surface Drainage of Pavements: Final Report. Washington, DC: The National Academies Press. doi: 10.17226/6357.
×
Page 40
Page 41
Suggested Citation:"Chapter 2 Literature Review and Current Practice and Techniques for Improved Surface Drainage." Transportation Research Board. 1998. Improved Surface Drainage of Pavements: Final Report. Washington, DC: The National Academies Press. doi: 10.17226/6357.
×
Page 41
Page 42
Suggested Citation:"Chapter 2 Literature Review and Current Practice and Techniques for Improved Surface Drainage." Transportation Research Board. 1998. Improved Surface Drainage of Pavements: Final Report. Washington, DC: The National Academies Press. doi: 10.17226/6357.
×
Page 42
Page 43
Suggested Citation:"Chapter 2 Literature Review and Current Practice and Techniques for Improved Surface Drainage." Transportation Research Board. 1998. Improved Surface Drainage of Pavements: Final Report. Washington, DC: The National Academies Press. doi: 10.17226/6357.
×
Page 43
Page 44
Suggested Citation:"Chapter 2 Literature Review and Current Practice and Techniques for Improved Surface Drainage." Transportation Research Board. 1998. Improved Surface Drainage of Pavements: Final Report. Washington, DC: The National Academies Press. doi: 10.17226/6357.
×
Page 44

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

CHAPTER 2 LITERATURE REVIEW AND CURRENT PRACTICE AND TECHNIQUES FOR IMPROVED SURFACE DRAINAGE Water films develop on the pavement surface during natural rainfall and tend to increase in thickness along the water drainage or flow path. At the onset of rainfall, the water first occupies the macrotexture on He pavement surface and is contained within the macrotexture of the pavement surface or is drained from Be surface through grooves or interns drainage (porous asphalt surfaces). With increasing rainfall, a film of water forms above He macrotexture. The flow of water on the pavement surface under these conditions is referred to as sheet flow, the depth of the sheet flow tends to increase in the direction of the drainage path. The kept of the sheet flow is of critical importance because the depth of this flow controls He skid resistance of He pavement and the tendency for hydroplaning. The vehicle speed at which hydroplaning occurs is inversely proportional to the depth of the sheet flow. The pavement design engineer must be able to identify any points on the pavement where sheet flow is sufficient to cause hydroplaning and must provide alternative or complementary strategies for reducing the kept of the water film Sickness. The models identified during this study provide He tools needed to calculate the depth of sheet flow as a function of four general pavement characteristics: pavement geometry, location and capacity of drainage appurtenances, surface texture of the pavement surface, and any internal drainage 11

offered by open-graded asphalt concrete (OGAC) surfaces or grooved Portland cement concrete pavements. The term open-graded asphalt concrete is used in this study to indicate either open-graded asphalt friction courses (OGAFC) or porous asphalt. Both types of mixes provide internal drainage; OGAFC is typical of U.S. practice, porous asphalt is typical of European practice. By varying any one or any combination of these characteristics, the Pavement design engineer can predict the effect of the characteristics on the water film thickness and, in turn, the propensity for hydroplaning. As part of this study, an interactive computer program, PAVDRN, was developed to predict water film thickness and tile potential for hydroplaning. The program is described in Appendix A. SI)MMARY OF MODELS NEEDED TO DEVELOP GUIDELINES The one~imension~, steady-state, kinematic wave equation was selected for calculating water film thickness in He computer-based design program, PAVDRN. The selection of a one-dimension~ flow equation was based on computations stability and efficiency. The major advantage of the one~imensional, kinematic wave mode} is that it is easy to apply and is computationally stable. A full description of this mode} is given in Chapter 3, where the development and rationale for choosing the various models used within PAVDRN are discussed. A number of other models were needed to develop PAVDRN. These include models for: 12

. . . Predicting the flow through porous pavement surface layers: For this purpose, the water film thickness mode] for impervious surfaces was modified to account for internal flow. Relating sight distance and vehicle speed to rainfall intensity i: a model from the AASHTO design guide (~) was selected for this purpose. · Predicting hydroplaning speed (HPS): A model first proposed by GalIaway (3) was used for this purpose. The HPS is a function of water film thickness and pavement macrotexture (MTD). Determining the hydraulic roughness coefficient, Manning's n: This is an empirical parameter (see equation 18) that depends on the type of surface and the Reynold's number, NR. The Reynold's number is a dimensionless parameter Hat is used to identify flow as laminar or turbulent (see equation 20~. Relationships were developed on the basis of data in He literature and new data collected as part of this study for three cases; Portland cement concrete pavements, dense-graded asphalt surfaces, and open-graded asphalt concrete surfaces. A full description of the rationale used in selecting these models and in their development is given in Chapter 3 and in Appendices B Trough D. METHODS FOR CONTROLLING WATER FILM THICKNESS A literature survey and a questionnaire were used to establish He current state~f-the art methods for pavement surface drainage. Implementable techniques for improving surface 13

drainage that resulted from He literature survey and from a questionnaire sent to 72 highway agencies can be grouped into four broad categories: · Optimization of geometric design parameters such as cross-sIope; · Reduction of the distance that the water must flow (flow path) by installing drainage appurtenances; Use of internally draining (asphalt concrete) wearing course mixtures; Use of grooving (per hard cement concrete); and · Maximization of surface texture. Controlling Water Film Thickness Through Pavement Geometry Highway geometric design criteria have evolved over many years and are designed to ensure He safe and efficient movement of vehicles. State agencies and many other transportation agencies use the guidelines issued by the American Association of Highway and Transportation Officials (AASHTO) (l) for geometric design. The drainage capacity of a highway surface is determined primarily by its surface geometry, especially cross-sIope. Geometric design criteria that enhance drainage are often in conflict with the design criteria for safety and driver comfort. Thus, although changes in the criteria contained in current geometric design guidelines may be desirable from the standpoint of improved drainage, there is little possibility that such changes will be effected solely for the sake of enhanced drainage. Geometric design criteria are presented in detail in He AASHTO design guidelines (~) but are 14

reviewed briefly here to illustrate geometric criteria that control surface drainage, but that must be satisfied during the pavement design process. The longitudinal slope of He pavement is referred to as its "grade." Criteria for both minimum and maximum grades are necessary for proper geometric design. Minimum allowable grades are necessary for drainage concerns, while maximum allowable grades must be specified for safety reasons and to control traffic flow. The longitudinal slope of the pavement and its surrounding gutters and ditches is usually the same within each section of highway. Therefore, this discussion covers all Free areas of the pavement system. Maximum grades have been established based on vehicle operating characteristics, particularly, the operating performance of larger vehicles such as tractor semitrailers. Steep grades can be difficult to descend and vehicles often reduce speed when ascending excessively steep grades. In general accordance with the AASHTO policy (l), maximum grades are determined by the functional class and design speed of the roadway and He surrounding topography. Typical maximum longitudinal grades are shown in table 1. The development of many of He models that were reported in the literature also was performed using regression in English units. The origins form of the models is retained throughout this report to maintain the integrity of the original analyses. Where English units occur, conversions between English and System International (SI) units are given in the text. Minimum grades are required to ensure adequate drainage. This is important for curbed roadways since water cannot drain laterally from a roadway when curbs are present. 15

Table I. Maximum recommended grades All. Design Speed mi/h (km/h) Design Section 30 (48) 40 (64) 50 (80) 60 (96) 70 (~12) Rural Sections, Maximum Grade, % Level -- 5 4 3 3 Rolling -- 6 5 4 4 Mountains -- 8 7 6 5 Urban Sections, Maximum Grade, % Level ~7 6 5 Rolling 9 8 7 6 Mountains ~ ~10 9 ~ The minimum grade recommended is 0.5 percent, but, if this cannot be obtained, 0.3 percent can be used as long as no curbs are present, and the roadway is crowned properly (]J. Vertical curves connect segments of constant grade. Since these curves often represent a change between a positive and negative grade, a level section exists at the transitions between positive and negative grades. AASHTO policy (]J is to design vertical curves using a ~K-value." The K-value is defined as Me horizontal distance in feet (meters) required to effect a I-percent change in the gradient of Me grade. To limit drainage problems, according to AASHTO All, K-values used in design should be less than or equal to 167 ft (51 m) for both crest and sag vertical curves. Basically, a K-value of 167 It (51 m) states that a 0.3-percent grade is the minimum grade allowed win 50 It (15 m) of Me level pavement on a vertical 16

curve. If K-values greater than 167 ft (51 m) are used, special attention should be given to the selection of pavement geometry to ensure adequate drainage. This is critical in sag vertical curves since water tends to collect at the bottom of these curves. Pavement cross-slopes (transverse) are a compromise between drainage (steep slopes) and driver comfort and safety (flat slopes). Cross-slopes may be formed in a number of ways, as shown in figure 4. In this figure, section 2 removes the water from the roadway faster than section 1, but more inlets are needed to collect He water at the edge of He pavement. These sections are recommended if freeze-thaw is common. Drainage can be directed in two ways, sloping toward the median or sloping toward the shoulder. If a highway slopes toward He median, more inlets will be needed, but less water will be in the outer travel lane, while more water will be in He inner, high-speed lane. Figure 4 shows a variety of drainage configurations including drains located within the traveled way. These are discussed in more detail in Chapter 2. This shows that cross-slopes are a compromise between many factors, and each has to be given serious consideration. As with longitudinal pavement slopes, a maximum and minimum superelevation is suggested by AASHTO (IJ. Research by Gallaway et al. (4) has shown that superelevations of two percent have lime effect on driver comfort or vehicle stability. The maximum transverse slope recommended by AASHTO is two percent per successive lane. The maximum slope permissible, as recommended by AASHTO (1), is four percent. Typical cross-slopes for various pavement types are presented in table 2. 17

J o a, · - o - a, co · - c) c', ·-- ID ~ 03 ~v i ~ 'ill ~ Or a) . ~ Q l,0~3G) 3 I ~ ~= = J .O O.c a' _ 0 ~ _ l F _ l cot 3= cat' 3 , C' ., see . 0 .= a) .. o a, ·' Cal ~ Figure 4. Different lateral drainage configurations with and without lateral drains. 18

Table 2. Typical cross-sIopes for different pavement surfaces Ail. Pavement Type Cross-slope - High Intermediate Low 1.5-2 1.5-3 2-6 On longitudinal curved sections of highways, the pavement is typically superelevated. A limit is placed on the rate of superelevation for driver comfort and safety. If the superelevation is too high and speeds are too low, drivers will need to steer up the slope. Also, vehicles can slide toward the inside of Me curve if ice is present on pavements with high superelevations. For reasons given above, the absolute maximum superelevation is 12 percent. Other maximum cross-slopes exist and are applied depending on the situation (]J. If ice anti snow are common, a maximum superelevation of eight percent is used. When heavy traffic volumes and low speeds prevail, a maximum slope of six percent is used. In urban areas, when speeds are low, curves can be designed without superelevation. A summary of maximum allowable values for superelevation rates is presented in table 3. 19

Table 3. Maximum allowable superelevation (1J. Situation Slope (%) Absolute Maximum Ice and Snow Uncommon Ice and Snow Common Urban, Low Speed 12 10 8 6 In order to obtain full superelevation on the curve from a tangent section of roadway, the n 1/3 rule" is commonly applied. This rule states that 2/3 of the superelevation should be obtained before the beginning of the horizontal curve. This transition distance is called the length of ~runoff." This runoff length can be obtained from most highway design manuals and is a function of design speed, highway curvature, and lane width. Transition is an important element in drainage design: When moving from a normally crowned pavement to a superelevated pavement, the pavement surface is usually rotated about the center line of the highway. This causes a section of Me pavement to be level. Consequently, when considering pavement surface drainage, special attention should be paid to these transition areas. After Me water has drained from the traveled lanes, the shoulders or parking lanes must either convey the water to an inlet or drain the water to ditches. Shoulders are typically used on rural roadways, while parking lanes and gutters are used in urban areas. Consideration of curbs, gutters, and other drainage appurtenances is beyond Me scope of this project; Hey are examined elsewhere (5,69. 20

Based on this research project, the authors recommend that AASHTO review its current policy on the geometric design of highways arm streets to consider establishing minimum cross-slope recor'?mendatior~s for highway pavements (1). The results of this study show that as the longitudinal slope or grade increases, the cross-slope of a pavement section should also be increased in order to remove water more rapidly from the pavement. This effectively shortens the distance a droplet of water must travel to reach He nearest appurtenance of a pavement edge (maximum flow path length; see figure 5), a critical design parameter for pavement drainage. In summary, the use of geometry to reduce water film thickness on pavements is constrained by the need to ensure driver comfort and vehicle stability. This effectively limits the maximum cross-slopes that can be used to remove water from the pavement, and thus other methods are required to enhance drainage and reduce He depth of water on the pavement. Most importantly, even though pavement geometry is an important factor in determining water film thicknesses, it alone may not correct drainage situations that lead to the potential for hydroplaning. Consequently, other means of drainage and water film Sickness control are needed as described in the following. Controlling Water Film Thickness Through Use of Appurtenances Drainage appurtenances are a very effective means for removing water and shortening the distance that water must flow in order to be removed from the pavement surface. Shortened flow paws imply reduced water film thickness. Traditionally, flow from He 21

Figure 5. Typical sloped drain. 22

pavement has been directed to the shoulder area and collected there. Drains installed between traveled lanes in the roadway surface itself have received little attention from design engineers or manufacturers of drainage appurtenances. Slotted drains in particular offer considerable potential as a means for shortening the length of the water flow path by simply reducing the distance that the water must flow before it is removed from the surface as illustrated in figure 4. Comprehensive analyses of Me interception capacity and spacing recommendations for drainage appurtenances have been performed in many studies (7,8). These studies have traditionally evaluated appurtenances located along the outer edge of the travel lanes. This section discusses the use of appurtenances located within the traveled section of the roadway between adjacent travel lanes. A questionnaire was sent to transportation agencies in August of 1993 as part of this project to identify current drainage practice. The responses indicate a general agreement among the agencies as to the preferred choice of methods for removing water from the pavement surface. The traditional procedure is to allow the water on multilane, high-speed highways to flow over the pavement surface to the shoulder(s). There, the water is channeled to a drainage swale or to a curb or gutter inlet. Depending on the geometry of the roadway section, appurtenances for collecting surface water can be placed on the outer edge of the travel lane or in the median section. Responses from the questionnaire indicate that the selection and spacing of curb opening inlets is usually determined in accordance with standard highway design guidelines such as the MSHTO Policy or Geometric Design of Highways and 23

Streets, Highway Drainage Guidelines published by AASHTO, Drainage of Highway Pavements: HEC-12, or individual agency standards (l, 6, 8). Responses from the questionnaire indicated that only seven agencies use slotted drains along the outer edge of the travel lane. One agency reported spacing slotted drains at intervals of 800 mm (2 It 8 in). Two agencies were considering using longitudinal slotted drains to drain curbed medians. Four states reported using longitudinal slotted drains between traffic lanes, and several state DOT's were considering their use. Slotted drains are pipe sections with an opening cut along the longitudinal axis and with transverse bars spaced in the opening to form slots, as shown in figure 5. Many configurations exist. They are produced by a number of manufacturers, and all manufacturers provide detailed descriptions of their drains as well as design criteria for their use. Although longitudinal slotted drains appear to be very attractive in terms of enhancing pavement drainage, slotted and other drains located within the traveled way do pose several potential problems that should be addressed. A possible disadvantage of such a system is the potential for plugging. In the event of plugging, severe ponding could develop on the pavement surface, creating a safety hazard. Retrofitting existing pavements to accommodate slotted drains within or between We traveled lanes would be costly except, during major rehabilitation if the pavement cross-slope must be modified to accommodate the drains. 24

Drainage structures within trafficked areas are subject to settlement resulting from traffic loads, causing unevenness in the roadway surface. Design procedures for supporting slotted drains when they are installed within the traveled (loaded) portion of the pavement need to be established. For this reason, Me installation of slotted drains at the edge of existing pavements may be the most cost-effective use of slotted drains. In summary, slotted drains are used only on a limited basis by highway agencies to drain the roadway surface. Their use by design agencies is encouraged. Placing longitudinal drains between traveled lanes is especially effective in reducing the flow path length (see figure 5), particularly for multi-lane pavements. Special consideration may be needed to provide structural support for drains within the traveled way; on the basis of this study, more widespread use of slotted drains is warranted, and studies held to implement the wider use of slotted drains should be initiated. The design of slotted as well as other drains and their capacity was beyond We scope of this study. Controlling Water Film Thickness with Internally Dmining Asphalt Surfaces (OGAC) Another technique for reducing water film Sicknesses on a roadway surface is We use of internally draining or open-graded asphalt concrete. The purpose of this discussion is not to research these asphalt surfaces per se, but to point out their potential use in minimizing water film thickness and hydroplaning potential. Therefore, a brief summary of the use of internally 25

draining asphalt concrete is given in this section. These surface mixtures can reduce the water film thickness; by (1) allowing internal drainage, which effectively reduces the amount of water that must be drained across the surface of the pavement and (2) by increasing the mew texture depth. Most research reports and engineers emphasize the internal drainage aspects of these mixtures, but the enhanced surface texture that Hey afford may be of equal or more importance Wan the interns drainage that Hey provide. The first use of porous or permeable surface layers in the United States occurred in the State of Oregon in the early 1930's (9). This pavement consisted of a surface treatment that was placed on an impermeable base layer. The permeable surface layer increased He frictional resistance of the surface, but He pavement was short-lived during periods of heavy traffic load. From this early work, open-graded asphalt friction courses (OGAFC) developed. These mixes typically contain 10 to 13 percent air voids (99 and are hot-laid with a paving machine to a depth of approximately 19 mm. The maximum aggregate size ranges from 13 to 19 mm. Asphalt content is selected as the maximum amount of asphalt that the hot mix can retain without appreciable drainage when the mixture is still hot. This is determined by placing mixes with differing asphalt contents on a plate in an oven and measuring the amount of asphalt Hat drains from the mix. These mixes offer increased skid resistance and allow interns drainage of surface water from He pavement surface (109. Open-graded mixtures with larger air-void contents, referred to as porous asphalt, drainage asphalt, or permeable asphalt, have evolved from He early use of open-graded 26

asphalt friction course OGAFC (9-1 7). These mixtures have been used extensively in Europe; they are placed in a thicker lift than OGAFC (usually greater than 25 mm thick) with binders that are modified win fiber or polymer (18919). These mixtures contain approximately 20 percent air voids, which is significantly higher than the OGAFC surface mixes used in the United States. Porous asphalt surfaces offer high values of skid resistance and contribute to the removal of water from the pavement surface. A summary of the mixture characteristics for different porous pavements as used in the United States and Europe is provided in table 4 (1~17920). The effectiveness of porous asphalt can be enhanced if drains are installed internally within the pavement layers. Continuous fabric drains that can be placed either transverse to or longitudinally with the direction of traffic have been used successfully for a number of years. The drains can be laid flat (drains have a rectangular cross-section) and may be placed with a new porous asphalt layer when the pavement is overlaid or during new construction. Details of this system and its use are given elsewhere (21). The use of porous asphalt pavements is a controversial subject with many state highway agencies. Although porous asphalt pavements are generally accepted as useful with respect to reducing hydroplaning, their performance has been unsatisfactory in many states, to the extent that several states have eliminated their use entirely. In contrast, they are used extensively on the motorways in Europe, especially in France and the Netherlands. By the year 2002, all of the motorways in the Netherlands must be surfaced win porous asphalt mixtures (221. 27

Table 4. Gradations used for internally draining asphalt mixes. Percent Passing Size Oregon (9) Typical Swiss Belgium France Europe (22) (13) (14) (17,]~) 25.0 mm 99-100 - - - - 19.0 85-96 100 - - - 14.0 - - - 100 100 12.5 60-71 - - 11.2 - 90-95 - 10.0 - - 100 9.5 8.0 - 28-40 6.3 17-31 - 5.0 4.75 2.75 2.36 2.0 18-23 7-19 10-12 710, ~- 6-8 250 - 4-6 90 - 2-4 74 1-6 - 17 5 55 23 14 - Air Voids (%) 5.7-10 17-22 14-20 16-28 24 Thickness (mm) 1.5-2.0 40-50 28-50 40 42 Permeabilitr, - 0.06~.12 0.057 0.0078- 0.02 (ifs) 0.023 28

The following are cited as advantages of porous asphalt pavements: I. Hydroplaning. Porous asphalt pavements reduce He thickness of the water film on He surface of He pavement, thus greatly reducing tendency for splash and spray from vehicles and the hydroplaning potential of the pavement. 2. Skid Resistance. The skid resistance for porous asphalt pavement is generally considered to be equal to that of traditional pavements. Testing performed by van der Zwan et al. (12J showed that at higher vehicle speeds, where aggregate macrotexture has a greater effect on skid resistance, porous pavement actually gives a higher skid resistance than conventional pavements. 3. Splash and Spray. Surface water can quickly infiltrate into porous asphalt, greatly reducing the amount of free surface water, which causes splash and spray from the vehicle tires. This reduction in splash and spray provides greater visibility, resulting in safer roadway conditions than on Portland cement concrete or conventional dense-graded asphalt pavements (23-25~. 4. Headlight Reflection. With the surface water infiltrating into the pavement, the reflections of vehicle headlights are greatly reduced and He visibility of roadway markings is increased. Porous asphalt surfaces offer a significant increase in surface texture over conventions dense-graded surfaces. The increased texture, in conjunction with internal drainage, can result in a significant reduction in the hydroplaning potential. However, Here are a number of disadvantages associated with these surfaces: I. Skid Resistance. At lower speeds, He skid resistance of porous asphalt is lower than for conventional asphalt surfaces, because Here is less aggregate surface at He 29

tire-pavement interface for porous asphalt mixes. The microtexture of these surfaces is generated primarily by the coarse-s~zed aggregate particles. This is not considered a serious disadvantage because on high speed motorways, skid resistance is critical at high speeds, not low speeds. 2. Plugging. There is a tendency for Me voids in porous asphalt surfaces to become plugged and filled with antiskid material and other roadway debris such as sediment runoff and material spilled on the road surface. During Weir first year of use, approximately one third of Me permeability of porous pavements is lost as a result of plugging (261. The French have concluded that a level of approximately 20 percent voids is needed for porous pavements to perform effectively. Therefore, current design practice in France requires initial void contents of 27 to 30 percent 3. Deicing Performance. Road salts tend to infiltrate into the surface voids reducing Me effectiveness of the salt or requiring larger application rates than for conventional surfaces. It takes three times the amount of salt on porous pavements as on traditional pavement types to produce the same deicing effects (my. Anti- skid materials also tend to plug the voids in porous pavements. 4. Black Ice. Porous asphalt surfaces have a tendency to develop black ice more quickly than conventional dense-graded pavements. Black ice can occur suddenly at the onset of a light rainfall when the internal pavement temperature is near or above freezing, and the air temperature is at or below freezing. Because porous asphalt conducts heat less readily than dense-graded mixtures the water on the pavement surface freezes more rapidly. The formation of black ice is a serious safety concern and has caused French authorities to discontinue the use of porous asphalt surfacings in Me Alps where Me conditions for Me formation of black ice are common. 30

5. Raveling. Raveling and loss of adhesion between porous asphalt surface layers and the underlying layers are He most frequently cited performance problems in the United States. However, the raveling problem may be alleviated by carefully selecting proper modifiers or the amount and type of asphalt binder in the mix (] 7- 199. 6. Delamination. There have been instances when the open or porous mixtures delaminated from the underlying pavement. This behavior, which occurred in Maryland in He winter of 1994, is apparently caused by the freezing action of water when the porous layer is saturated (2i7. The cause of the delamination is hypothesized as follows: If the freezing of the water in the layer proceeds simultaneously from the top and the bottom simultaneously, there is no outlet for the expanding water as it freezes. The expanding water then creates sufficient force to delaminate the surface layer. Extensive delamination caused Maryland to abandon open-graded mixtures. There appears to be a general consensus among pavement engineers that porous asphalt surfaces can greatly reduce the potential for hydroplaning. Porous asphalt surfaces also reduce tire noise and minimize splash and spray, thereby increasing driver visibility (11,18,25,26,28). Reducing splash and spray makes He roadway safer to Ravel on during periods of rainfall. Tappeiner (20) cites a European report that states that there were 20 percent fewer fatalities and injuries by motorists while traveling on porous asphalt pavements during wet weaker conditions. A similar reduction was also reported in the United States (91. These claims for improved safety must be considered within the context of the French experience where problems with black ice formation have been observed. In reviewing the advantages and disadvantages of porous asphalt surfaces, it can be seen that this type of pavement has many positive attributes if careful attention is given to mix proportions, materials selection, and construction details. Porous asphalt surfaces, especially 31

newer mixture designs with special binders, warrant greater use in the United States although their disadvantages must also be carefully considered. Condoning Water Film Thickness with Grooving The fours method for reducing water film thicknesses is the use of grooving on Portland cement concrete surfaces. Grooving is generally ineffective on asphalt concrete surfaces because the grooves close quickly under the action of traffic. The grooves in Portland cement concrete act as subsurface channels that drain water from the pavement surface. The use of grooving for airport pavement has considerable attention by researchers (29J. Typical grooving patterns used for airport runways are shown in figure 6 (30). A typical airport runway may consist of two 3~m (100 fit) wide lanes, each sloping laterally from a center crown at 1.5 percent. Hence, grooves are perpendicular to the wheel path and in the direction of the water flow. To be fully effective, the grooves should be parallel to the direction of flow; for highways with both a longitudinal and across slope, the grooves must be skewed to the direction of traffic if the grooves are to be parallel to the water flow. This is often not practical and reduces the effectiveness of Me grooves as drainage channels. Reed et al. (30J used dye tests to confirm that all the rain falling on the upstream end of the flow path was carried in the grooves in such a way that the slab surface, although wet, had zero water film thickness. The width of the area contributing lateral inflow to the grooves was equal to the groove spacing for spacings of 127 mm (5 in) or less. It was also possible to predict the location where the grooves began to overflow, overflowing water contributes to the water film thickness. The down-slope point at which the grooves were full, and runoff began to spill out onto the pavement surface was called the breakout point. This point serves as the origin for sheet flow and the point where water film starts to develop. 32

it-- Sinai Inflow Is :~1 ~6.4muTI / ~ ~ ~ ///////////// \ I i/" ! 6.4 ( a ) Rectangular groove pattern s = x, 127 mm, 64 mall 32 null. ~70m ~21nun~ =~ 77 ( b ) Refle~c-percussive groove pawed. 6.4 mm Figure 6. Typical grooving patterns for Portland cement concrete pavement (30). 33

The breakout point was computed by considering the equilibrium flow rate and the capacity of the grooves, bow of which were a function of rainfall rate and down-sIope distance. The equation for the breakout point is (311: L, (1.50) / (s i ng) (1) where Breakout distance measured from top edge of pavement (ft)(1 It - 3.05 m) s = Groove spacing (in)(1 in = 25.4 mm) i - Rainfall rate (in/h)(1 in/in = 25.4 mm/in) n. Manning roughness coefficient for grooves . g The coefficient 1.50 is a function of groove geometry, and as given in equation 1 is for 6-mm- by-6-mm (0.25in-by~.25-in) rectangular grooves with a pavement surface slope equal to 1.5 percent. The results of data generated by Reed et al. t30) for grooved Portland cement concrete pavements are summarized in figure 7. The figure was developed for a rainfall intensity of 75 mm/in (3 in/h). The breakout points are shown on the graph as the intersection of the curves with the abscissa. The smaller the groove spacing, the greater is the distance to the breakout point L. The Marlning roughness coefficient for the grooves, ns, was taken as 0.01 (31). Grooving can reduce the water film Sickness on pavements by acting as drainage channels and thereby carrying water from the pavement surface. However, unless grooves are parallel to the slope of the pavement, their ability to conduct flow is reduced and their effectiveness minimized. In summary, grooving Portland cement concrete pavements can reduce water film thickness and thus increase the speed at which hydroplaning will occur. This has been demonstrated for grooves whose principal orientation is in Me direction of the flow paw of the water (309. The PAVDRN model, documented in Appendix A, uses 34

~ \ i\ ; \' to to to O I' 8 ° z" C, I' _ _ U' _ ~ == 3 ~ ·= Z ~ ~ U) \ \ \ \ \ ~\ \ -\ \ - C~ I_ - _ O c: A: ~ _ _ ~ · A X HI ·C~ Ct ~ co En at at O O O O C!) C' ~ E Cal CD ~ - ., .\ \ 1 ' 1 1 ''. \'' ~d . . \ ·C~ Q V) o ~ 2 ~ o ~ \ o ~ Z \ I . ; \ c., ; \ \ es ;, ~\ se `. Cal - ~n Cal . . . ~ 1~.-.--,,---.--~-, , ,+ , , ~. - ~C9 ~ ~ C ~C ~O U]UJ '4IdeG JaleM - ce Q o o C~ CD - cn ._ Figure 7. Predicted water film thickness, WFr, for grooved portland cement concrete pavement, rainfall iritensity 75 mm/in (309. 3s

information about groove spacing, width, and depth to effectively increase the mean texture depth of Me pavement and thus increase the speed at which hydroplaning occurs. ControNing Water Film Thickness with Surface Texture Another method for controlling water film thickness is by maximizing the texture of the pavement surface. The water film thickness is the total thickness of the film of water on the pavement minus the water trapped in the macrotexture of the pavement surface. Water film thickness is reduced in direct proportion to the increase in macrotexture (total macrotexture volume, not MID). The importance of macrotexture for asphalt surfaces is discussed in a previous section on the use of porous asphalt to control water film thickness. Since porous asphalt surfaces are typically prepared from relatively coarse aggregates or gradations with a minimal quantity of sand-sized material, they generally yield large levels of macrotexture. The macrotexture of other asphalt surfaces is also controlled by the gradation of the aggregate, ranging from very low levels of macrotexture for sand asphalt to relatively large levels of macrotexture for coarse-graded mixture and surface treatments. The importance of macrotexture is recognized in French practice where microsurfacing techniques are now widely used and have replaced porous asphalt in areas where the performance of porous asphalt has been suspect (see also the section on porous asphalt as a method for controlling water film thickness). Micros urfaces are Tin lifts of hot-mix asphalt concrete graded to maximize surface texture. Macrotexture is also important for Portland cement concrete surfaces. New Portland cement concrete pavement surfaces in the United States are typically constructed with fined surfaces to enhance macrotexture. Macrotexture produced by fining or brooming is to be distinguished from grooving. The texture of Portland cement concrete pavement can be enhanced by etching away the mortar exposing the coarse aggregate (new construction) or by grinding (to restore texture in old pavements) although these techniques are not used often in practice and often result in high levels of tire noise. 36

The importance of texture is recognized in the reproposed Design Guidelines for Improving Pavement Surface Drainage" (2) where the pavement texture is one of the design options. The importance of macrotexture may not always be demonstrated with the standard ASTM E 274, "Standard Test for Skid Resistance of Paved Surfaces Using a Full-Scale Tired locked wheel tests because the test is not conducted on a flooded pavement. Instead, the water introduced in front of the tire of the moving skid tester. For example, in full-scale tests conducted at the Turner-Fairbanks Highway Research Center, no correlation was shown between macrotexture and hydroplaning speed (32J. PROPOSED DESIGN GUIDELINES FOR IMPROVING PAVEMENT DRAINAGE- IMPLEMENTATION OF FIN1)INGS The design guidelines were developed as a ~stand-alone" document for use by design engineers in the design of new roadway systems or the rehabilitation of existing pavements (2). The guidelines can be used by highway design engineers to evaluate the effect of different pavement parameters on the water film thickness and the potential for hydroplaning. The guidelines are complemented by an interactive computer program, PAVDRN, which allows the pavement engineer to predict water film thickness and the propensity for hydroplaning (Appendix A). The treatment of the different design parameters is reviewed briefly in this section. The reader is referred to the Proposed Design Guidelines for Improving Pavement Surface Drainage" (2) for details. Pavement Geometry Five different types of design sections are considered in the guidelines and in the PAVDRN computer program. They include (~) tangent sections, (2) superelevated horizontal curves, (3) transition sections, (4) vertical crest curves, and (5) vertical sag curves. Each of these sections can be analyzed using the PAVDRN model. In the analysis, the pavement is divided into successive sections or planes according to one of the five types of design section 37

(see Appendix A, especially figure A-2). The flow from one type of section to another can be linked in the analysis. Geometric information is required for each section in the analysis. For tangent sections, the guidelines recommend that, as grade increases, pavement cross-slope should also be increased up to maximum recommended values. The guidelines also recommend other control methods such as slotted drains between traveled lanes. For superelevated sections the guidelines suggest the use of a maximum recommended superelevation to minimize water film thickness on horizontal curves and the use of other methods, such as increased mean texture depth or grooving, if superelevation does not reduce the potential for hydroplaning to desired levels. In transition sections, the effects of changes in the pavement geometry on the flow path length are fairly complex. The location of the maximum flow path length changes depending upon the difference between the cross-slope at the curve end of the transition and the cross- slope at the tangent end. Runout length also affects the location of the flow path and its length. The runout length is the distance, measured from the start of the plane and in the direction of the traveled way, to the point where the Towpath exits He plane. In general, the guidelines recommend that the runout length be shortened as cross-slopes increase. However, in transition sections concern for safety and driver comfort must be balanced. Other measures to control water film thickness might need to be applied if the shortest recommended runout length is used, and He potential for hydroplaning still exists. Pavement Properties There are two pavement-related factors that can be controlled by the designer to control the water film thickness: (1) pavement type and (2) mean texture Kept. Four pavement types are considered in the guidelines and the PAVDRN software. The four pavement types are: (1) Portland cement concrete (PCC), (2) grooved PCC (GPCC), (3) dense-graded asphalt concrete 38

(DGAC), and (4) OGAC, which includes both porous asphalt and open-graded asphalt friction course (OGAFC). The design information required to specify the design section varies with He pavement type. In PAVDRN, the mean texture depth (MTD) is a function of several parameters that are determined by the designer. For PCC surfaces, the water-to-cement ratio and the surface finish (e.g. degree of lining) affect the MTD. Maximum aggregate size, gradation, and air void content affect the texture of asphalt concrete mixes. OGAFC and porous asphalt surfaces have larger macrotexture than dense-graded surfaces. Porous mixtures and high air-void content mixtures both contribute to the mean texture depth and, in An, to a reduction in the water film Sickness. Grooving PCC pavements reduces water film thickness if the grooves are oriented so that they conduct flow off the pavement. Otherwise, the effect of grooving is localized and can lead to increased water film thickness on other parts of the pavement. The guidelines provide specific recommendations win respect to groove size and spacing based upon an analysis using PAVDRN and survey responses from highway engineers. Drainage Appurtenances The designers ability to reduce water film thickness on a highway pavement using geometry and pavement properties is limited. Drainage appurtenances are typically necessary to control water film thickness, especially on large, multilane facilities where the flow path length spans more Man two travel lanes. The most promising technology for multilane highways is Be use of slotted drains placed between the Gavel lanes. At least four state transportation departments reported using slotted drains in this manner. Slotted drains can also be placed transversely or across the traffic lane to capture flow. Drains used in either manner reduce the water film thickness on a pavement by removing or reducing flow over the pavement. 39

PAVDRN SOFIA PAVDRN is intended for use by highway design engineers to determine the likelihood of hydroplaning on various highway pavement sections. It does this by computing the longest flow path length over the design pavement section and determining the water film thickness (depth of water above the asperities of the pavement surface) at points along the path. The water film thickness is used to estimate the speed at which hydroplaning will occur (if at all) along the longest flow path, the critical path in the section. The predicted hydroplaning speed along this path is then compared to the design speed of the facility, a parameter selected by the designer. PAVDRN nuns under Windows 3. ~ and above. The user interface was programmed in Visual Basic. The computational algorithms were programmed in FORTRAN 77. The user interface uses point-and~lick technology with pull-down menus and context-sensitive help. The user's guide is available on-line and is installed as part of the help files with the PAVDRN program. Since it is a one-dimension model' PAVDRN first analyzes the section geometry to determine the maximum or longest flow path length over the pavement section. The program determines water depth, time to equilibrium, and velocity at points along the longest flow path length; equations for determining these values are presented in Chapter 3. The mean texture depth is subtracted from He depth to determine the water film Sickness. The water film thickness, computed in this manner, is used to determine the speed at which hydroplaning will occur. Results are printed in a summary report format. They are also available as a text file that can be imported to a third-party graphics program and plotted. A sample of the summary output table is provided in table 5 based on the analysis of a tangent section with zero grade and standard I.S-percent cross-slope. 40

Table 5. PAVDRN summary output table. X Y Di stance OFT Flow/width Manning' I; n Reynold' s No. Hydr. Speed (ft,m) (ft,m)(ft,m) (in,mm) (cfs/ft,cms/m)(mi/h) 150.0.0 .00 -.SOE+OO .00E+00 .000 O. 999999 150.01.0 1.00 .64E+00 .14E-04 .310 11. 109 150.02.0 2.00 .88E+00 .28E-04 .076 21. 100 150.03.0 3.00 .lOE+01 .42E-04 .061 32. 96* 150.04.0 4.00 .12E+01 .56E-04 .0S2 42. 93* 150.05.0 5.00 .13~+01 .69E-04 .046 53. 91* * Indicates hydroplaning speed is; ~ es`; than design "peed . Note: PAVDRN ha" the option of producing output in SI or English unit=. The data shown in the table abjure are in U.S. unit". Design Ex~nple Using SIoHed Drains ' This analysis considers a tangent section consisting of three lanes of Gavel in Me same direction. The geometric input for the analysis of ache tangent section in this example is listed in table 6. Table 6. Tangent section properties. Property Value No. of Planes Length of Each Plane Longitudinal Slope Width of Each Plane Pavement Type Mean Texture Depth Cross-Slope of Plane ~ Cross-SIope of Plane 2 Cross-Slope of Plane 3 3 300 m 0.02 m/m 4m PCC 0.50 mm 0.015 m/m 0.025 m/m 0.035 m/m 41

Additionally, a rainfall intensity of 80 mm/in is assumed, and a kinematic viscosity of the water of 1.306 x 10 ~6m2/s (water temperature = 10 °C) was chosen. These values of intensity and water temperature are conservative but might be observed in some locations in the United States. A summary of the output of the model is shown In table 7. Table 7. PAVDRN output for tangent section. Water Film End of Drainage Length Thickness Flow/Widd ~Hydroplaning Plane (m) (mm) (m3/s/m) Speed (krn/h) 1 6.66 1.3 0.00013 90 2 11.79 1.5 0.00023 88 3 16.39 1.6 0.00032 86 The results In table 7 present the final value of the water fihn thickness at the end of the longest drainage path length across each section of the pavement. In this example, since each lane has a different cross-slope a plane consists of one lame of travel. At the end of the first plane, the model has predicted that the flow length of water across the innermost lane will be 6.66 mm, and the hydroplaning speed at that point is 90 km/in. The lane is only 4 m wide, but the flow length will be along a distance that is the resultant of the cross-slope and the longitudinal slope. Therefore, the drainage path length will be greater than the 4-m width. For a design speed of 90 kn'/h, the computed hydroplaning speed just meets the criteria to prevent hydroplaning. However, as the drainage length increases across the second and third lanes of travel, the water film thickness increases to a point where the hydroplaning speed for the third, outermost lane of travel is significantly below the design speed of 90 1an/h. One solution for increasing the hydroplaning speed could be to install a longitudinal slotted drain between the second and third lanes of travel In the direction of travel (see figure 5~. This drain 42

would intercept the flow from Me second lane, reduce the water film thickness at the end of the second lane, and additionally reduce We water film thickness across the entire third lane of travel. This would reduce the hydroplaning potential of the entire roadway system to be in accordance with a design speed of 90 km/in. From the analysis, ache flow at the end of the second plane needs to be reduced to a value that will eliminate the hydroplaning potential for the system. A slotted vane grate is selected and placed between the second and third lanes. Using a design chart provided by the manufacturer to obtain a grate inlet coefficient, K = 39, the grate will capture (0.000516 m3/s per meter of length of the slotted drain inlet) as determined by equation 2: Q = K D5/3 where Q D Flow rate (cfs/ft) Depth of flow (ft) ( lcfs - 0.028 m3/s, 1 It = 305 mm) At this location in the pavement, the flow is only 0.00023 m31sim, and therefore the total flow will be captured. 43 (2)

To PAGE INTENTIONALLY LEFT BLANK

Next: Chapter 3 Selection and Development of Models for PAVDRN »
Improved Surface Drainage of Pavements: Final Report Get This Book
×
MyNAP members save 10% online.
Login or Register to save!
  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  6. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  7. ×

    View our suggested citation for this chapter.

    « Back Next »
  8. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!